Answers are grounded in this paper and cite their sources.
Rainfall-induced landslides threaten the Wumeng Mountains Region (WMR), while projected population decline and ageing may alter who remains exposed. We integrated an extreme gradient boosting (XGBoost) susceptibility model trained with 2,941 rainfall-induced landslide samples and ten conditioning factors with a performance-selected NEX-GDDP-CMIP6 ensemble, SSP-consistent land-use projections, and population and socioeconomic scenarios. XGBoost achieved a testing ROC-AUC of 0.794. Relative to 1995–2014, mean annual precipitation and rainfall-threshold exceedance frequency increase by 2.73%–15.92% and 3.24%–18.19%, respectively, across the assessed future periods and scenarios. The proportion of very-high-susceptibility area increases from 3.91% historically to 12.98% in late-century under SSP5-8.5, with 46.83% of the region shifting to a higher susceptibility class. Although the exposed population generally declines, the population exposure rate remains 40%–47%; children and older adults exceed 50% of exposed people in late-century under SSP1-2.6 and SSP5-8.5, and exposure rates in relatively low-income areas remain 38.87%–50.81%. Thus, declining exposed population does not necessarily imply a proportional reduction in relative exposure burden. The framework identifies where increasing susceptibility overlaps with persistent demographic and socioeconomic exposure, supporting targeted adaptation in population-declining mountain regions.
No takes yet. Share an insight, caveat, or question.
Huang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: